ABSTRACT NUMBER: ESOC2026LB136 LB136 - PERSONALIZED PREHOSPITAL TRIAGE FOR SUSPECTED ACUTE STROKE (PRESTO-2): IMPLEMENTATION AND EVALUATION OF A REAL-TIME DECISION SUPPORT TOOL

Abstract Background and aims Prehospital triage of suspected stroke patients aims to minimize treatment delays by avoiding interhospital transfers for patients with ischemic stroke due to large vessel occlusion (LVO) eligible for endovascular thrombectomy (EVT). Current triage tools such as prehospital stroke scales rely on fixed cut-points and largely ignore patient- and system-level heterogeneity. We implemented and evaluated a novel real-time decision support tool providing personalized prehospital triage advice by integrating stroke severity, onset time, GPS-based driving times, hospital-specific workflow metrics, and time-dependent treatment effects. Methods We developed the Stroke Triage App, a CE-registered software-based medical device used by paramedics. The Stroke Triage App provides personalized triage advice by comparing the predicted probability of good functional outcome (mRS 0-2) for transport to the nearest primary stroke center versus direct transport to a more distant thrombectomy-capable center. The app was implemented in two ambulance regions in the Netherlands covering approximately 1.8 million inhabitants. We performed a prospective before-and-after evaluation with a concurrent control region using a difference-in-differences linear regression analysis. The primary outcome was the difference in onset-to-groin time, a key determinant of functional outcome. Results Between January 22, 2024 and December 31, 2025, 2,687 patients were triaged using the Stroke Triage App across the two intervention regions. Data collection and analyses are ongoing. Final results will be presented at ESOC 2026. Conclusions Implementation of personalized prehospital decision support with the Stroke Triage App has the potential to reduce onset-to-groin times in patients with LVO ischemic stroke. Conflict of interest